In aplastic anemia, the bone marrow is replaced by fat, and peripheral blood counts - - of white blood cells, red blood cells, and platelets - - fall to extremely low levels, leading to death from anemia, bleeding or infection. Aplastic anemia is a disease of young persons, and in its severe form is almost invariably fatal untreated. Historically, aplastic anemia has been linked to chemical exposures, in particular benzene; it is an idiosyncratic complication of some medical drug use; it occurs as a rare event in pregnancy and following seronegative hepatitis; and the diseases associated with certain immunologic conditions. The chance observation that some patients post-bone marrow transplant recovered their own marrow function led to the inference that the immunosuppressive conditioning regimen might have treated an underlying immune-mediated pathophysiology. Purposeful administration of antithymocyte globulin (ATG) has led to hematologic recovery in the majority of treated patients. Laboratory data have also revealed abnormalities of the immune system: lymphocyte populations that induce apoptosis in hematopoietic target cells by the Fas-mediated pathway, and oligoclones of effector T cells which express type 1 cytokines, especially gamma-interferon. The Hematology Branch has been a leader in both the scientific and medical studies of aplastic anemia pathophysiology and treatment. Several major clinical protocols have been completed or are in progress. We have determined that prolonging cyclosporine administration from the standard six month period to two years post-ATG delays but does not prevent relapse. Approximately the same 1/3 of patients will eventually require further immunosuppression, whether or not additional cyclosporine is prescribed. However, analysis of the data suggests that most relapse occurs at relatively low doses of cyclosporine, and future protocols might aim to treat patients at 1-2 ml/kg beyond six months; this dose is unlikely to result in serious toxicities. We have continued our studies of eltrombopag, a thrombopoietin mimetic which can be administered orally and is approved for use in refractory idiopathic thrombocytopenic purpura. We published last year that 40-50% a small cohort of patients with refractory severe aplastic anemia showed hematologic responses to eltrombopag. As these responses were robust, occurred in trilineages, and were accompanied by increased bone marrow cellularity, eltrombopag was hypothesized to act as a stem cell factor. The refractory study has been extended, with additional patients showing approximately the same proportion and pattern of response. Further, we have enrolled almost two dozen patients in a novel new protocol, which combines standard immunosuppression with horse ATG and cyclosporine with eltrombopag in severe treatment nave disease. Preliminary data from this ongoing trial suggests that the response rate may be higher than the usual 65%, and that responses are occurring earlier and are more robust, leading to a higher proportion of patients who are free of transfusion as soon as one month after ATG and who have achieved almost complete responses at three months. In this study, ancillary laboratory assays show a marked increase in CD34 cell number in bone marrow, consistent with the stem cell stimulation mechanism. However, in both our studies of refractory and treatment-nave disease, the major concerns remains stimulation of abhorrent clones, which bear cytogenetic abnormalities and may be associated with refractory cytopenias, myelodysplasia, and acute myeloid leukemia. Other current studies of eltrombopag in the Branch include treatment of patients with low risk myelodysplastic syndrome and with moderate aplastic anemia. We have completed translational studies related to our landmark publication showing that horse ATG is superior to rabbit ATG. We utilized samples from the 120 patients treated in this trial to assess differences in the immune response to these heterologous protein preparations, as well as correlates of serum sickness, for which ATG treatment is a good model in humans. We found that rabbit ATG was detectable in the blood for a much longer period than was horse ATG, and was bound to lymphocytes in the circulation. Neutrophil numbers were much lower in rabbit ATG-treated patients. Rabbit ATG also resulted in the production of multiple cytokines detectable in the plasma. While rabbit ATG induced higher frequencies of certain lymphocyte subsets, its depletion effect on CD4 cells overwhelmed enhancement of T-regulatory cell development. Plasma rabbit ATG levels predicted the occurrence of serum sickness, and its prevalence peaked around week two, concurrent with the clearance of ATG from blood and the production of a variety of cytokines. Horse ATG and rabbit ATG therefore have very different pharmacokinetics as well as effects on cell subsets and cytokines, differences likely related to both their efficacy and toxicity. In studies in the mouse, we have expanded on earlier descriptions on a model for an immune-mediated bone marrow failure based on infusion of lymph node cells different in either major H2 or minor histocompatability loci. We have developed a model for marrow failure in B6 mice utilizing infusion of FVB lymph node cells, which appears to create a more chronic model of marrow depression, allowing investigation of treatments after pancytopenia develops, thus more closely mimicking the human disease. In a separate project, we have investigated the possibility that adipocytes are important factors in producing hematopoietic depression. However, in contrast to prominently published work, we only observed this correlation in the setting of immune-mediated bone marrow failure. Inhibition of adipogenesis with specific chemical agents was effective in improving bone cellularity and blood counts in our aplastic anemia model but not after transplantation, or in radiation or chemotherapy induced bone marrow failure. The explanation for this discrepancy is that the inhibitors of adipogenesis, PPAR &#947; agents, also act on T cells to suppress their function. These experiments depend on the specificity of PPAR gamma and PPAR antagonists and their specificity. In addition to inhibiting adipogenesis, as determined by micro arrays, the purportedly specific PPAR &#947; agonist reduced CD8 and CD4 T cell infiltration in the bone marrow and inflammatory cytokine levels in plasma. In addition, inflammasome genes were also decreased as determined by microarrays.